Indian Journal of Urology
: 2018  |  Volume : 34  |  Issue : 1  |  Page : 39--44

Oncologic outcomes in patients with nonurothelial bladder cancer

Sanjay G Patel1, Adam Benjamin Weiner2, Kirk Keegan3, Todd Morgan4,  
1 Department of Urology, University of Oklahoma Health Sciences Center, Oklahoma City, OK, USA
2 Department of Urology, Northwestern University, Chicago, IL, USA, India
3 Department of Urologic Surgery, Vanderbilt University Medical Center, Nashville, TN, USA
4 Department of Urology, University of Michigan Medical School, Ann Arbor, MI, USA

Correspondence Address:
Sanjay G Patel
Department of Urology, University of Oklahoma Health Sciences Center, Oklahoma City, OK


Introduction: We aimed to evaluate the relative prognostic impact of the most common variant histologies on disease-specific survival (DSS) in patients undergoing radical cystectomy. Materials and Methods: The Surveillance, Epidemiology, and End Result database was used to identify patients who underwent radical cystectomy for bladder cancer from 1990 to 2007. Patients with urothelial cell carcinoma (UCC), squamous cell carcinoma (SCC), adenocarcinoma (AC), sarcoma, small cell carcinoma, signet ring carcinoma, and spindle cell carcinoma were included in the study. Multivariable analysis was performed using Cox proportional hazards model to assess independent predictors of disease-specific survival (DSS). Mortality rates were estimated using Kaplan–Meier analyses. Results: A total of 14,130 patients met inclusion criteria with the following histologies: UCC (90.1%), SCC (4.6%), AC, (2.3%), sarcoma (0.8%), small cell carcinoma (0.8%), signet ring carcinoma (0.5%), and spindle cell carcinoma (0.9%). Three-year DSS was most favorable in patients with UCC (63.7%; 95% confidence interval [62.9%–64.8%]) and AC (65.3% [59.3%–70.6%]), whereas 3-year DSS was the least favorable for small cell carcinoma (41.6% [31.3%–51.6%]) and sarcoma (45.4% [35.1%–55.1%]). In the multivariable analysis, independent predictors of DSS were age, marital status, grade, T-stage, N-stage, and variant histology. With respect to UCC, there was an increased risk of disease-specific death associated with all variants except AC. Sarcoma and spindle cell carcinoma were associated with the highest risk of death. Conclusions: With the exception of AC, the most common variant bladder cancer histologies are all independently associated with worse DSS relative to UCC in patients undergoing radical cystectomy.

How to cite this article:
Patel SG, Weiner AB, Keegan K, Morgan T. Oncologic outcomes in patients with nonurothelial bladder cancer.Indian J Urol 2018;34:39-44

How to cite this URL:
Patel SG, Weiner AB, Keegan K, Morgan T. Oncologic outcomes in patients with nonurothelial bladder cancer. Indian J Urol [serial online] 2018 [cited 2022 Oct 2 ];34:39-44
Available from:

Full Text


Bladder cancer accounts for over 70,000 new cancer diagnoses per year in the United States, with pure urothelial cell carcinoma (UCC) comprising approximately 90%–95% of these cases.[1],[2],[3] The remaining 5%–10% of bladder cancers consist of pure nonurothelial histologies or mixed urothelial and nonurothelial histologies, and these are generally associated with a worse prognosis compared to urothelial cell bladder carcinoma.[4] Given the paucity of information regarding nonurothelial histologies, there is a tendency among clinicians to homogenize all the clinically and biologically different pure nonurothelial and mixed histologies into a single category: nonurothelial. Improved understanding of relative survival patterns among pure nonurothelial histologies is needed to improve patient counseling and aid clinical decision-making before cystectomy.

Due to their rarity, institutional series of nonurothelial bladder carcinoma has generally evaluated only a single histology at a time and is unable to place each histology within the context of other histologic subtypes.[4] Studies involving larger administrative databases have been performed as well, but only select nonurothelial histologies have been assessed.[5],[6],[7]

No comparative analyses assessing the multiple variant histologies and their impact on oncologic outcomes have been performed. We, therefore, sought to evaluate the relative prognostic impact of the most common variant histologies on long-term survival in patients undergoing radial cystectomy using the Surveillance, Epidemiology, and End Results (SEER) database.

 Materials and Methods

We reviewed the SEER database, the countries' preeminent cancer database (, from 1990 to 2007 for patients diagnosed with bladder cancer. To homogenize our cohort to patients who were considered surgical candidates, we limited our study population to patients who underwent radical cystectomy. Patients who were under 18 years of age or presented with metastatic disease were excluded from the study.

Organ site was extracted using International Statistical Classifications of Diseases for Oncology, 3rd edition (ICD-O-3) site codes for the urinary bladder (C670–C679). Histology ICD-O-3 codes were used to identify the most commonly observed bladder cancer histology cell types: UCC (8120, 8130), squamous cell carcinoma (SCC) (8070, 8076), adenocarcinoma (AC) (8140, 8144, 8255, 8260, 8263, 8310, 8323, 8480, 8481, 8570, 8574, 8575), sarcoma (8801-8802, 8810, 8890, 8896, 8880), small cell carcinoma (8041, 8446), signet ring carcinoma (8490), and spindle cell carcinoma (8030, 8031, 8122).

Covariates included demographic factors such as age, sex, race, year of diagnosis, geographic location, and marital status. Pathologic factors included pT stage, pN stage, and tumor grade. Demographic and pathologic factors were compared using Chi-squared and Fisher's exact test where appropriate for categorical variables and t-test continuous variables. The primary outcome was disease-specific survival (DSS) and the secondary outcome was overall survival (OS). Survival was calculated from the date of diagnosis to the date of last follow-up or death. If death was not the final outcome, patients were censored at the date of the last follow-up. To compare the survival among different histologic subtypes, mortality rates were estimated using Kaplan–Meier product limit method and compared using the log-rank test. Univariable and multivariable analyses were performed using the Cox proportional hazards model analysis. Potential confounding variables included in the multivariate analysis were selected based on P < 0.10 in univariate analysis or clinical appropriateness.[8] Statistical analysis was performed using Stata version 13.0 (College Station, TX, USA). All reported P values generated were two-sided and P < 0.05 was considered statistically significant.


A total of 14,130 patients with the following histologies: UCC (90.1%), SCC (4.6%), AC (2.3%), sarcoma (0.8%), small cell carcinoma (0.8%), signet ring carcinoma (0.5%), and spindle cell carcinoma (0.9%) were reviewed. The median follow-up time was 25 months (interquartile range 10–61 months).

The demographic and pathologic characteristics among the different histologic subtypes are shown in [Table 1]. The percentage of each patient population comprised males ranged from 80.3% of patients with signet ring tumors and 55.7% of patients with SCC. Only AC was more likely to occur in non caucasian patients compared to UCC (9.9% vs. 20.5%, P < 0.001).{Table 1}

Only spindle cell had similar T-stage compared to UCC (P = 0.057) and only sarcoma had similar nodal status (P = 0.114). The rate of nodal metastases ranged from 10.7% in sarcoma to 27.5% in signet ring.

The univariable hazard ratio (HR) and 3-year and median OS and DSS by histologic subtype is shown in [Table 2]. With the exception of AC, all nonurothelial bladder cancers have a worse 3-year OS and DSS when compared to UCC. The worst DSS was demonstrated in small cell histology with 3-year survival probabilities of 41.6%. [Figure 1] and [Figure 2] show Kaplan–Meier survival curves for DSS and OS stratified by histology, respectively. The curves for UCC and AC never reach their median DSS within 10 years of diagnosis.{Table 2}{Figure 1}{Figure 2}

Multivariable survival analyses were performed controlling for age, grade, and pT and pN stage are shown in [Table 3]. Being married was associated with improved DSS (HR 0.84, 95% confidence interval [CI]: 0.79–0.90, P= 0.001). Compared to urothelial carcinoma, there was an increased risk of disease-specific death associated with all variants except AC (HR 1.01, 95% CI: 0.82–1.23, P= 0.92) and small cell (HR 1.40, 95% CI: 1.00–1.97, P= 0.49). Sarcoma was associated with the worst DSS (HR 1.85, 95% CI: 1.36–2.51, P < 0.001), whereas signet ring tumors had the worst OS (HR 1.70, 95% CI: 1.25–2.32, P= 0.001).{Table 3}


While UCC is the most common histology for patients undergoing radical cystectomy, histologic variants comprise a substantial number of patients undergoing radical cystectomy in the United States. Using a large administrative data set, we report an incidence of nonurothelial cancer of approximately 10% in patients undergoing radical surgery. This is high relative to previous studies which report rates ranging from 5% to 10%.[1],[2] The present study is the first to perform a comparative survival analysis for multiple histologic variants relative to urothelial carcinoma while controlling for age and other adverse pathologic characteristics. Survival information for patients with rare bladder tumors underrepresented in the literature provided by our study will prove useful in counseling newly diagnosed patients.

Similar to patients with pure UCC, patients with variant histology bladder cancer tended to be males although the patient population with SCC was nearly equal in terms of gender composition. Similarities in nonwhite race patient populations were also noted with Whites comprising near 90% of most populations; however, AC had nearly twice the nonwhite population compared to urothelial carcinoma (20% vs. 10%, respectively).

AC had similar OS and DSS to UCC. Ploeg et al. demonstrated a similar finding in an analysis of the nationwide Netherlands Cancer Registry in all patients with >T1 disease where patients with urothelial carcinoma and AC had similar relative 3-year survival rates of 37% and 39%, respectively.[9] We show that this similarity holds true even when the study population is limited to patients who underwent radical cystectomy. The improved survival among patients with AC compared to other nonurothelial bladder cancer is multifactorial and includes less aggressive tumor biology and tendency for local not distant recurrence.[10] In a series of 192 patients with bladder AC treated with cystectomy alone, 79 had disease recurrence, over half of which was local recurrence.[10] Despite these local recurrences, patients with pT1-2a disease had 100% disease-free survival 5 years following radical surgery and those with pT2b and pT3 had 47% and 40% disease-free survival at 5 years, respectively.[10]

Small cell carcinoma of the bladder also had similar overall and DSS compared to urothelial carcinoma on multivariable analysis despite having worse 3-year survival on log-rank test. This finding may be secondary to the small number of patients with small cell carcinoma in our data set. A recent case–control study matched 68 patients with small cell carcinoma of the bladder with 138 patients with urothelial carcinoma of the bladder with similar TNM stage who received radical cystectomy.[11] This study also noted no difference in survival although adjuvant chemotherapy improved survival. Multimodal therapy, surgery and a combination of radiation therapy and chemotherapy, has emerged as the predominant treatment for patients with small cell carcinoma of the bladder.[7] While there is no standard treatment guideline, data from the National Cancer Database and MD Anderson Cancer Center showed that neoadjuvant chemotherapy greatly improved OS and DSS compared to radical cystectomy alone.[7]

Besides AC and small cell carcinoma, the remaining analyzed histologic variants were associated with worse OS and DSS following radical cystectomy. The reasons for the worse survival in these patients are certainly multifactorial. First, the tumor biology and natural history of nonurothelial tumors may have more aggressive characteristics and thus portend a poor prognosis. Second, the paucity of published clinical series is in general fragmented with series limited to institutional series makes optimal treatment regimens and timely management of these tumors difficult. Finally, discordancy between transurethral resection of bladder tumor (TURBT) specimens diagnosed at pure urothelial carcinoma and final radical cystectomy histologic designation of mixed histology may preclude certain patients from receiving potential neoadjuvant multimodal therapy. In one multicenter study, more than half men with variant histology noted upon radical cystectomy specimen analysis were not diagnosed with variant histology at TURBT.[12]

We also present the largest cohort of patients with signet ring tumors who underwent radical cystectomy. In a recent review of case reports of 54 patients, only 8 patients survived >2 years.[13] Of these, all but one received radical cystectomy.[13] Radical cystectomy is generally considered the treatment of choice for localized signet ring carcinoma.[14] Platinum-based drugs, 5-fluoriouracil, and anthracyclines are generally selected as chemotherapeutic agents but are not consistently effective.[13] Thus, our survival data on patients with localized signet ring tumors of the bladder who underwent radical cystectomy represent accurate reflections of the likely poor prognosis for a patient with these tumors.

Differences in survival may be related to lower utilization of chemotherapy for variant histologies, for which there is no consensus regimen. The role of neoadjuvant and adjuvant chemotherapy in patients with variant histology has received some recent attention but remains poorly understood. In a secondary analysis of the SWOG 8710 study which evaluated survival in patients with stage T2-T4a bladder urothelial carcinoma undergoing cystectomy versus cystectomy plus neoadjuvant methotrexate, vinblastine, doxorubicin, and cisplatin chemotherapy, there was a survival benefit in the patients with mixed histology (urothelial carcinoma + squamous and glandular differentiation) (HR 0.46; 95% CI: 0.25–0.87; P= 0.02).[15] While this analysis was performed on mixed tumors, there may be some benefits for neoadjuvant therapy if extrapolated to the treatment of pure nonurothelial bladder carcinoma. Black et al. in a series from MD Anderson Cancer Center advocate the use of neoadjuvant chemotherapy in high-risk urothelial carcinoma and consider nonurothelial histology a high-risk feature and thus a qualifier for neoadjuvant chemotherapy.[4]

As a retrospective study utilizing administrative data, our study has important limitations. The SEER data set contains information that is retrospectively collected and relies on the ability of individual coders to translate information from patient charts into the data set. There is no information regarding certain relevant covariates such as patient comorbidity, margin status, indication for RC, prior therapies, or adjuvant or neoadjuvant chemotherapy administration, which could contribute to a bias in survival analysis. Furthermore, there was no central pathology review, which may have resulted in misclassification of these rare and often difficult to characterize bladder histologies. Finally, it is impossible to determine whether histologic variant occurred in the presence of absence of coexisting UCC.

Despite these limitations, the power of the SEER data set lies in its large sample of these relatively infrequent bladder cancers over a large population that is broadly generalizable across the United States. The rarity of these nonurothelial tumors makes it difficult to obtain sufficient sample sizes from institutional series; therefore, administrative data sets can offer an important avenue to more thoroughly understanding the relative significance of variant histologic subtypes of bladder cancer.[16] In addition, SEER provides valuable information on DSS, information not available in the National Cancer Database or the Netherlands Cancer Registry.[7],[9]


With the exception of AC, most common nonurothelial bladder cancer histologies are all independently associated with worse DSS relative to UCC in patients undergoing RC. In particular, sarcoma, spindle cell, and signet ring variants were particularly aggressive malignancies, more so even than small cell carcinoma of the bladder. Further research is needed to better understand the unique biology of these nonurothelial bladder cancer histologies as well as the optimal timing and modes of therapy to improve the long-term outcomes in affected patients.

Financial support and sponsorship:


Conflicts of interest:

There are no conflicts of interest.


1Dahm P, Gschwend JE. Malignant non-urothelial neoplasms of the urinary bladder: A review. Eur Urol 2003;44:672-81.
2Kantor AF, Hartge P, Hoover RN, Fraumeni JF Jr. Epidemiological characteristics of squamous cell carcinoma and adenocarcinoma of the bladder. Cancer Res 1988;48:3853-5.
3Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA Cancer J Clin 2015;65:5-29.
4Black PC, Brown GA, Dinney CP. The impact of variant histology on the outcome of bladder cancer treated with curative intent. Urol Oncol 2009;27:3-7.
5Abdollah F, Sun M, Jeldres C, Schmitges J, Thuret R, Djahangirian O, et al. Survival after radical cystectomy of non-bilharzial squamous cell carcinoma vs urothelial carcinoma: A competing-risks analysis. BJU Int 2012;109:564-9.
6Wright JL, Porter MP, Li CI, Lange PH, Lin DW. Differences in survival among patients with urachal and nonurachal adenocarcinomas of the bladder. Cancer 2006;107:721-8.
7Patel SG, Stimson CJ, Zaid HB, Resnick MJ, Cookson MS, Barocas DA, et al. Locoregional small cell carcinoma of the bladder: Clinical characteristics and treatment patterns. J Urol 2014;191:329-34.
8Bursac Z, Gauss CH, Williams DK, Hosmer DW. Purposeful selection of variables in logistic regression. Source Code Biol Med 2008;3:17.
9Ploeg M, Aben KK, Hulsbergen-van de Kaa CA, Schoenberg MP, Witjes JA, Kiemeney LA. Clinical epidemiology of nonurothelial bladder cancer: Analysis of the Netherlands Cancer Registry. J Urol 2010;183:915-20.
10Zaghloul MS, Nouh A, Nazmy M, Ramzy S, Zaghloul AS, Sedira MA, et al. Long-term results of primary adenocarcinoma of the urinary bladder: A report on 192 patients. Urol Oncol 2006;24:13-20.
11Kaushik D, Frank I, Boorjian SA, Cheville JC, Eisenberg MS, Thapa P, et al. Long-term results of radical cystectomy and role of adjuvant chemotherapy for small cell carcinoma of the bladder. Int J Urol 2015;22:549-54.
12Cai T, Tiscione D, Verze P, Pomara G, Racioppi M, Nesi G, et al. Concordance and clinical significance of uncommon variants of bladder urothelial carcinoma in transurethral resection and radical cystectomy specimens. Urology 2014;84:1141-6.
13Akamatsu S, Takahashi A, Ito M, Ogura K. Primary signet-ring cell carcinoma of the urinary bladder. Urology 2010;75:615-8.
14Fiter L, Gimeno F, Martin L, Gómez Tejeda L. Signet-ring cell adenocarcinomaof bladder. Urology. 1993;41:30-3.
15Gore JL, Kwan L, Saigal CS, Litwin MS. Marriage and mortality in bladder carcinoma. Cancer 2005;104:1188-94.
16Rogers CG, Palapattu GS, Shariat SF, Karakiewicz PI, Bastian PJ, Lotan Y, et al. Clinical outcomes following radical cystectomy for primary nontransitional cell carcinoma of the bladder compared to transitional cell carcinoma of the bladder. J Urol 2006;175:2048-53.